Create a project on Azure Devops Server

2. Go to create free pipleine

3. Create an organization

   

4. Create a project

Create build pipeline

1. Go to pipeline, and click on new pipeline

2. Choose github as source code repository

   

3. After Oauth with GitHub, choose the repository (saxo-university)

4. Choose the starter pipeline for configuration options

   

5. Now we can edit teh starter pipeline boilerplate

   

6. Edit the pipeline yam to create three task (as scripts)

• yarn install

• yarn lint

• yarn build

  # This sets trigger to any update in master  
  # trigger:
  # - master
  
  #  Defines the host for agent (MS hosted)
  pool:
    vmImage: 'ubuntu-latest'
  
  #  Build steps for the pipeline
  steps:
  - script: yarn 
    displayName: 'Installing node modules'
  
  - script: yarn lint
    displayName: 'Checking for lint errors'
  
  - script: |
      yarn build
    displayName: 'Running webpack build'
  
  # Save artifacts to refer them in release builds
  - task: PublishPipelineArtifact@1
    inputs:
      path: $(Agent.BuildDirectory)/s/build
      artifact: react-static-website-build-artifacts

11. Select the yaml path in repository, this will be saved in repo on commit

    

Define work directory for all steps : https://github.com/MicrosoftDocs/vsts-docs/issues/6315

12. Click on save and run, choose to commit changes in master branch

Publishing artifacts

• To be able to deploy the bundled resources (e.g. to Azure Storage), we will need to retain and share the results (files) of this build with release build

• Lets instruct the azure to save the result of our build aka Pipeline Artifacts

• Add following task to the pipeline.yml

      # Save artifacts to refer them in release builds
      - task: PublishPipelineArtifact@1
        inputs:
          path: $(Agent.BuildDirectory)/s/projects/react-static-website/build
          artifact: react-static-website-build-artifacts

• Now after the build is run, open the build result, we will see an option to view/download artifacts :

  

Create release pipeline

• Go to release tab and click on New Pipeline

• 

• Click on Artifacts | + Add to select artifacts that will be deployed in this build

• Choose the build from dropdown to get the artifacts from

  

• Click on add and now we will see the build pipeline as :

  

Create release task to deploy to the dev environment

• Click on Stages | +Add

• Choose empty job and click on Apply

  

• Choose name of the stage and click on close button of stage dialogue

  

• Click on '1 job, 0 task' to add a task to the stage

  

• Click the + icon on the agent job to create a task

  

• Search for "Azure file copy" for available azure tasks:

  

• Click on newly created task File Copy to start editing it

  

• Edit the task as

  • Display name as Deploy to Dev Server

• Click on ... button next to Source input, this will show a dialog to select available artifacts

  • select the react-static-website-build-artifacts (we create this in our build )

• make sure you select the right dir here.

  

• Choose free trial from available Azure subscription dropdown

  

• Click on authorize to allow release task to connect to the Azure Storage service, using our free trial subscription

  • this will open and OAuth window with microsoft live

  

• Select Destination Type from dropdown as staticwebsiteazure (our dev server)

• In container name, enter $web

• 

• Click on save button to save the create build task

  

Configure Auto Deployment of artifacts from master branch

• Go back to pipelines tab to continue editing the release pipeline

• Click on trigger icon over the artifacts:

  

• Click on enabled in the dialogue

  

• Add filter to make sure the continuous deploy only happens for master branch

• click on **+ Add ** button below the Build branch filters and enter master in branch name

• now click on close on top of dialog

  

• Click on the Save on top right to save the new trigger

Auto deploy is ready

Now make a commit to the repository and it should auto deploy

Create production release with approval

• Every release to master gets deployed on the dev envrinemnt, however with production we need more control

• For production we will follow this workflow

  • create a release for master on dev
  • test on dev, if everything works fine, then we approve the artifacts to be delpoyed to production
  • only selected memebers should have the privilege to be able to approve a production release